cross_validation_xgbclassifier.py

from sklearn.model_selection import cross_validate
from sklearn.model_selection import StratifiedKFold

def print_scores(cv):
    means = np.mean(list(cv.values()), axis=1)
    [print(f"\tAverage {x[0].strip('test_'):<10} over all folds: {x[1]:.2f}") for x in zip(cv.keys(), means) if "test_" in x[0]]
    print()

cv = {}
metrics = ["accuracy", "precision", "recall", "f1", "roc_auc"]
for key in ["HashingVectorizer", "TfidfVectorizer"]:
    xgb_model = XGBClassifier(n_estimators=100, use_label_encoder=False, eval_metric="logloss")
    skf = StratifiedKFold(n_splits=5, random_state=42, shuffle=True)
    cv[key] = cross_validate(xgb_model, X[key], y, cv=skf, scoring=metrics)
    
    print(f"{key}:")
    print_scores(cv[key])